CN109858797B - Multi-dimensional informatics analysis method based on knowledge network accurate online education system - Google Patents
Multi-dimensional informatics analysis method based on knowledge network accurate online education system Download PDFInfo
- Publication number
- CN109858797B CN109858797B CN201910073607.5A CN201910073607A CN109858797B CN 109858797 B CN109858797 B CN 109858797B CN 201910073607 A CN201910073607 A CN 201910073607A CN 109858797 B CN109858797 B CN 109858797B
- Authority
- CN
- China
- Prior art keywords
- learning
- evaluation
- service object
- knowledge
- outline
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of online education emotion analysis, and relates to a multi-dimensional informatics emotion analysis method based on a knowledge network accurate online education system. The method predicts the future learning effect of the service object based on the full learning trace, the learning effect and the comprehensive evaluation, so that the service object can master the curve of the change of the learning effect in advance and track the learning data of the service object at any time; recommending an individualized teaching scheme by using a neural network algorithm, wherein the recommended individualized teaching scheme has pertinence; according to the comprehensive evaluation index clustering and dynamic identification, a teacher can arrange differentiated teaching measures according to different service object groups, and the teaching concept of teaching according to the situation is embodied.
Description
Technical Field
The invention belongs to the technical field of online education plot analysis, and particularly relates to a multi-dimensional informatics plot analysis method based on a knowledge network accurate online education system.
Background
With the popularization and continuous development of computer network technology, the internet access threshold is continuously reduced. With the popularization of intelligent terminals and the rapid development and maturity of matching technologies thereof, and the continuous improvement of requirements of students and parents on learning conditions, learning methods and learning tools, online education is accepted by more and more parents and students. More and more students select the online learning system to assist in improving learning of themselves, and the learning results of themselves are evaluated and checked on the internet.
The existing mainstream online education system generally gives the correct rate and wrong questions of students simply according to the evaluation and answer conditions of the students, most systems cannot well guide the individual learning of the students according to the multi-dimensional information such as the learning conditions of all the students, teaching outline, learning traces of individual students and the like, and predict the learning effect of the students, so that the students and teachers can only passively adjust the learning plan and the teaching scheme after the learning situation changes; these systems also generally lack a teaching scheme for customizing individuation according to comprehensive evaluation and learning traces of students, and fail to well implement education ideas of teaching according to the situation.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a multi-dimensional informatics analysis method based on a knowledge network accurate online education system, which is based on knowledge network accurate online education, performs multi-dimensional comprehensive learning evaluation on a service object around the learning trace of the service object and multi-dimensional emotional information of the service object, recommends an individualized teaching scheme, provides a reinforced learning plan, expands the individualized teaching scheme and optimizes an individualized learning promotion scheme.
The invention is realized by adopting the following technical scheme:
the multidimensional informatics analysis method based on the knowledge network accurate online education system comprises the following steps:
constructing a knowledge network accurate online education system;
based on a knowledge network accurate online education system, recording multi-dimensional learning situation information according to learning traces of service objects;
forming multi-dimensional comprehensive learning evaluation of the service object by utilizing a knowledge network emotion comprehensive evaluation method based on the existing learning trace and multi-dimensional emotion information of the service object;
recommending an individualized teaching scheme according to a teaching outline and multi-dimensional comprehensive learning evaluation of a service object;
recording new learning traces and evaluation results generated by the service object learning according to the recommended personalized teaching scheme, and evaluating the personalized learning effect of the service object through a multidimensional comprehensive learning evaluation index;
judging the category of the service object according to the full learning trace and the multidimensional comprehensive learning evaluation index of the service object, proposing a reinforcement learning plan, expanding a personalized teaching scheme and optimizing a personalized learning promotion scheme;
and according to the learning trace and the multidimensional learning situation information generated by the service object, a neural network model is adopted, and an individualized teaching scheme recommendation model is iteratively optimized.
Further, the knowledge network accurate online education system comprises:
(1) A knowledge network management engine;
(2) A knowledge network formed based on the correlation relationship of the knowledge elements;
(3) A knowledge network resource subsystem;
(4) A user management subsystem;
(5) A service object personalized evaluation subsystem;
(6) A schema management subsystem;
(7) A service object learning effect evaluation subsystem;
(8) A personalized teaching scheme recommendation subsystem;
(9) A clustering analysis and identification subsystem;
(10) And a student emotion analysis subsystem.
Further, the knowledge network management engine is responsible for management of the knowledge elements, and the functions comprise: (1) adding, deleting and modifying knowledge points; (2) labeling a knowledge element; (3) storing the knowledge points and the knowledge elements; (4) constructing a knowledge network based on the relation of the knowledge elements; (5) establishing learning evaluation resources based on the knowledge network, the knowledge elements and the corresponding knowledge resources; (6) and recording and storing the learning trace.
Further, various data generated when the service object uses the knowledge network accurate online education system to learn are recorded in the learning trace, wherein the various data comprise learning time, learning duration, the breadth and depth covered by the knowledge elements learned at a certain time, learning resource selection preference, learning behaviors, learning speed and progress, evaluation arrangement, evaluation method and evaluation result information.
Preferably, the multidimensional integrated learning evaluation is evaluated by evaluation indexes including: outline completion, intellectual competitiveness, adherence, concentration, self-expansion and practice.
Further, the multidimensional comprehensive learning evaluation method comprises the following steps:
(1) the adherence is measured by the study input time, the study ending time, the study input frequency and the study input time; the adherence force calculating method is to finish learning according to the study input time recorded in the study traceTime, learning frequency, and learning time duration, calculating the variance with the current learning time, learning frequency, and learning time duration to obtain variance vector D = (D) 1 ，d 2 ，d 3 ，d 4 ) A weight vector W = (W) from four elements 1 ，w 2 ，w 3 ，w 4 ) The final retention was obtained as:
(2) the intellectual competitiveness is measured by age, learning speed, evaluation effect, and the breadth and depth of learning covering knowledge elements; u = (U) 1 ，u 2 ，u 3 ，u 4 ，u 5 ) Respectively representing age, learning speed, evaluation effect, learning coverage knowledge point breadth and learning coverage knowledge point depth; evaluation scale V = (V) 1 ，v 2 ，v 3 ，v 4 ，v 5 ) Poor, qualified, good and excellent, respectively; after the service object finishes learning the knowledge element, the knowledge network accurate online education system combines U and V to give a scoring matrix R; obtaining a fuzzy vector B = A × R through the weight vector A, and introducing a score vector S, so that the intelligence competitiveness I:
I＝B*S T
wherein, the weight vector and the score vector are given by knowledge network experts or by an analytic hierarchy process;
(3) the self-expansion capability is measured by the comprehensive capability of the evaluation service object in the existing knowledge coverage range and the capability of expanding and solving the problem of directly subsequent related knowledge elements;
(4) practical ability P ra The evaluation service object is obtained by applying the obtained knowledge to solve the actual problem, and the capacity of the expansion problem or the practical capacity generated by a single expansion problem or practical problem is divided into the following steps according to the solution score s, the problem difficulty d and the time t for solving the problem:
wherein: k is a radical of 1 、k 2 Is a constant;
(5) concentrated force F f Measured by the learning activity time ratio:
wherein: t is t t Is the distraction time, t is the total learning time; the distraction time consists of the time when the page recorded in the service object full learning trace exceeds the pause limit, the page cut-out time and the page extra rolling time;
(6) the outline completion condition is measured by an outline requirement completion ratio, namely, the ratio of the completion degree of the outline after learning of a service object to the completion degree required by the outline as a whole after accumulating the products of the item-by-item completion degree of the outline and the importance degree of the outline is obtained by taking a knowledge element as a unit, and the outline completion condition of the knowledge element is calculated in the following way:
wherein: d i Importance of the ith outline, p i The completion degree of the ith outline service object is defined, and A is the completion degree required by the outline;
the above single self-expansion capability, single practice capability, single concentration force and single outline completion condition are obtained by adopting an iterative updating mode, wherein the overall expansion capability, practice capability, concentration force and outline completion condition is as follows:
wherein i = {1,2,3,4} represents self-expansion ability, practical ability, concentration ability, and outline completion, respectively, and d i Numerical value n representing a single index i Representing the number of evaluations of a single index, T, at the start of an iteration i ＝0；
And (3) overall multidimensional comprehensive learning evaluation:
E stu ＝(I，P，C d ，F，E，P ra )W T
wherein, I, P and C d 、F、E、P ra Respectively intelligence competitiveness, insistence, outline completion, concentration, self-development ability and practice ability, W = (W) 1 ，w 2 ，w 3 ，w 4 ，w 5 ，w 6 ) For the weight, the weight W is obtained by an expert or by an analytic hierarchy process.
Further, the evaluation result is divided into an actual evaluation result and an expected evaluation result, wherein the actual evaluation result is calculated in the following way:
wherein G is i A score representing the ith service object;
the expected evaluation result is calculated in the following manner:
wherein s is i The score of the ith question; theta is the service object learning ability; p is i (theta) the accuracy of the answer to the ith question of the service object, wherein the accuracy is calculated by the following method:
wherein: d is constant, D =1.7; a is the discrimination of the test questions; b is the difficulty of the test question; c is a guess parameter which represents the probability that the service object can still answer the pair when the comprehensive evaluation of the service object is negative infinity.
Further, the step of judging the category to which the service object belongs is given by using a fuzzy clustering analysis method and a fuzzy pattern recognition method, and comprises the following steps:
according to the relation among all indexes in the multidimensional comprehensive learning evaluation, the service objects are divided into a plurality of types through a fuzzy clustering analysis method, a system or a teacher can customize a teaching scheme individually according to the types of the service objects, a teacher and the system can conveniently recommend resources and adjust teaching according to the types of the service objects, and the types can be dynamically changed according to the full learning trace of the service objects;
and judging the category of the service object by using a fuzzy pattern recognition method according to the type of the service object and the multi-dimensional comprehensive learning evaluation index.
Preferably, the optimization-based personalized teaching scheme recommendation model is based on the goodness of the recommended teaching scheme, and the goodness evaluation of the recommended teaching scheme is measured by the variance of the expected learning effect and the actual learning effect.
Further, the service object is a student or a learning organization, the learning organization including a class, a school, a learning group, and/or an interest group.
The invention predicts the future learning effect of the service object through the past full learning trace and learning effect of the service object and comprehensive evaluation, and recommends the personalized teaching scheme according to the prediction result, compared with the prior art, the beneficial effects obtained by the invention include:
1. based on the past full learning trace, the learning effect and the future learning effect of the comprehensive evaluation and prediction service object, the service object can master the curve of the change of the learning effect in advance and track the learning data of the service object at any time, so that the learning state, the time arrangement, the learning plan and the like are adjusted in advance.
2. Aiming at the changes of the learning effect and the comprehensive evaluation of the service object, a neural network algorithm is used for recommending an individualized teaching scheme, and the recommended individualized teaching scheme has pertinence; according to dynamic clustering and dynamic identification of comprehensive evaluation indexes, a teacher can arrange differentiated teaching measures according to different service object groups, and the teaching concept of teaching according to the factors is embodied.
3. The accurate online education system of knowledge network can be according to the actual study trace and the recommendation result self-iteration of service object, self-renewal, reach higher rate of accuracy, it is more reliable.
Drawings
FIG. 1 is a flow chart of a mathematical plot analysis according to an embodiment of the present invention;
FIG. 2 is an illustration of an educational system in accordance with an embodiment of the present invention;
FIG. 3 is an abstract diagram of knowledge point associations for a knowledge network, in accordance with an embodiment of the invention;
FIG. 4 is a diagram of learning traces in accordance with an embodiment of the present invention;
FIG. 5 is a schematic diagram of a personalized instruction scheme according to an embodiment of the invention;
FIG. 6 is a graph of a comprehensive evaluation of six-dimensional tension in one embodiment of the present invention;
FIG. 7 is a diagram illustrating fuzzy clustering according to an embodiment of the present invention.
Detailed Description
In order to further illustrate the technical means and the technical effects of the present invention adopted to solve the technical problems, the following clearly and completely describes the technical solutions of the present invention with reference to the accompanying drawings and examples, but the embodiments of the present invention are not limited thereto.
The invention provides a multidimensional informatics analysis method based on a knowledge network accurate online education system, and aims to comprehensively learn and evaluate the personalized learning and learning effects of a service object based on the knowledge network accurate online education system around teaching resources, service objects and full learning traces recorded by the whole system, evaluate six basic quality indexes (outline completion, intelligent competitiveness, concentration, insistence, self-expanding capability and practical capability), predict the change of the comprehensive evaluation indexes by combining the full learning traces and the comprehensive evaluation indexes and provide a path, a method and teaching resources for learning improvement and recommend a personalized teaching scheme by using a neural network algorithm.
As shown in fig. 1, in the present invention, the accurate online education system based on the knowledge network records multidimensional learning information (the multidimensional learning information includes the learning time, the learning duration, the breadth and depth of the coverage of the knowledge elements learned at a certain time, the learning resource selection preference, the learning behavior, the learning speed and progress, the evaluation arrangement, the evaluation method, and the evaluation result) according to the learning trace of the service object (the learning trace records the behavior of the service object when learning in the accurate online education system based on the knowledge network and the generated intermediate data), forms the personalized comprehensive learning evaluation of the service object, constructs the personalized education recommendation scheme, evaluates the personalized learning effect of the service object, and optimizes the personalized teaching scheme recommendation model. The service object may be a student or a learning organization including, but not limited to, a class, a school, a learning group, or an interest group. The multidimensional learning context information refers to information which is used for analyzing the learning context of the service object after the learning trace of the service object is processed by the learning context analysis subsystem. A learning purpose of the knowledge network accurate online education system is shown in figure 2, and learning behaviors of students comprise recording learning tracks, calculating comprehensive evaluation, analyzing learning effects and recommending teaching schemes.
A multi-dimensional informatics analysis method based on a knowledge network accurate online education system is shown in figures 1-7 and comprises the following steps:
s1, constructing a knowledge network accurate online education system;
in this embodiment, the accurate online education system of knowledge network includes:
(1) A knowledge network management engine;
the knowledge network management engine is responsible for the management of the knowledge elements, and the functions include but are not limited to: (1) adding, deleting and modifying knowledge points; (2) labeling a knowledge element; (3) storing the knowledge points and the knowledge elements; (4) constructing a knowledge network based on the relation of the knowledge elements; (5) establishing learning evaluation resources based on the knowledge network, the knowledge elements and the corresponding knowledge resources; (6) and recording and storing learning traces.
(2) A knowledge network formed based on the correlation relationship of the knowledge elements;
a knowledge element in the system can be a knowledge point, a knowledge subnet or a knowledge point cluster, wherein: the knowledge subnet comprises knowledge points contained in all learning paths related by taking the knowledge point set as a starting point or an end point and relations among the knowledge points; the knowledge point cluster is composed of a series of knowledge points which cannot form a directly connected knowledge subnet; the relationships between knowledge elements are shown in fig. 3, and include a predecessor relationship, a successor relationship, a parent-child relationship, and a parallel relationship (also referred to as "sibling relationship"). The knowledge points are connected with the knowledge points related to the knowledge points according to the relationship set by the system to form a three-dimensional knowledge network.
(3) A knowledge network resource subsystem;
and the knowledge network resource subsystem manages knowledge resources and learning evaluation resources based on the knowledge elements.
In this embodiment, the learning and evaluation resources are an exercise question bank and a test question bank which are constructed by using the knowledge elements, the comprehensive tests and the like as basic test units.
(4) A user management subsystem;
(5) A service object personalized evaluation subsystem;
the personalized evaluation subsystem tracks the learning condition of the service object and dynamically forms personalized evaluation according to needs by utilizing a personalized evaluation system.
The types of the test questions of the personalized evaluation system comprise example questions, example promotion questions, memory type questions, application questions, calculation questions, comprehensive questions, expansion questions and the like, and are mathematically expressed as T = (T) 1 ,t 2 ,t 3 ,t 4 ,...,t M )，t i And the number of the test question types is not less than 0,1,M. The score for service object j is calculated as:
wherein, g i Is the fraction of the ith type question; s ji The score of the service object j in the test of the i types of questions is obtained; k is the number of categories of the type topic.
(6) A schema management subsystem;
the outline management subsystem manages the teaching outline based on the teaching target.
(7) A service object learning effect evaluation subsystem;
the learning effect evaluation subsystem tracks the learning condition of the service object and dynamically forms personalized evaluation according to needs by using the personalized evaluation system; the service object learning process and the evaluation result form an individualized learning trace and are stored in the accurate online education system, and the evaluation subsystem evaluates the learning effect of the service object according to the individualized learning trace of the service object.
(8) A personalized teaching scheme recommendation subsystem;
the personalized teaching scheme recommendation subsystem recommends a personalized teaching recommendation scheme covering a series of knowledge elements to the service object based on knowledge resources and evaluation resources, learning ability of the teaching outline and the service object and personalized requirements based on the knowledge network and according to the teaching outline and learning purpose, learning interest, learning ability and teaching and learning progress of the service object. The personalized teaching recommendation scheme points to a learning guidance scheme recommended by the service object, and the learning guidance scheme comprises learning resources, learning paths and the like, and the service object can reach an ideal learning situation state through the guidance scheme.
(9) A clustering analysis and identification subsystem;
(10) And a student emotion analysis subsystem.
S2, recording multi-dimensional learning situation information according to learning traces of service objects based on the knowledge network accurate online education system;
the knowledge network accurate online education system records the learning condition of a service object in the system and acquires multi-dimensional learning trace information. The learning trace information is original data which is not processed by the system, and is provided for other functions in the system after being subjected to operation processing such as normalization, sequencing and the like by the system. The learning traces are shown in fig. 4, and record various data generated when the service object uses the knowledge network accurate online education system to learn, including learning time, learning duration, breadth and depth covered by knowledge elements learned at a certain time, learning resource selection preference, learning behavior, learning speed and progress, evaluation arrangement, evaluation method, evaluation result and other information.
S3, forming multi-dimensional comprehensive learning evaluation of the service object by utilizing a knowledge network learning emotion comprehensive evaluation method based on the existing learning trace and multi-dimensional learning emotion information of the service object;
and forming multi-dimensional comprehensive learning evaluation on the current learning condition of the service object according to the breadth and depth of the knowledge elements covered by the service object learning recorded in the learning trace, the investment duration, the learning speed and progress and the evaluation effect.
The multidimensional comprehensive learning evaluation is evaluated through evaluation indexes, wherein the evaluation indexes include but are not limited to outline completion, intelligence competitiveness, insights, concentration, self-expanding capability and practical capability. The accurate online education system comprehensively evaluates the change trend of each index and analyzes the reason of the change.
In this embodiment, the multidimensional comprehensive learning evaluation is comprehensively evaluated by six indexes, namely, outline completion, intellectual competitiveness, insistence, concentration, self-expansion capability and practical capability. The specific quantitative method for evaluating the multidimensional comprehensive learning and forming the six indexes of the multidimensional comprehensive learning evaluation is a knowledge network theory comprehensive evaluation method, and specifically comprises the following steps:
(1) the adherence is measured by the study input time, the study ending time, the study input frequency and the study input time. The calculation method of the adherence force is to calculate the variance with the current study input time, study end time, study frequency and study input time according to the study input time, study end time, study frequency and study input time recorded in the study trace to obtain the variance vector D = (D) 1 ，d 2 ，d 3 ，d 4 ) A weight vector W = (W) from four elements 1 ，w 2 ，w 3 ，w 4 ) The final retention was obtained as:
(2) the intellectual competitiveness is measured by age, learning speed, evaluation effect, and the breadth and depth of learning covering knowledge elements. Let U = (U) 1 ，u 2 ，u 3 ，u 4 ，u 5 ) Respectively representing age, learning speed, evaluation effect, learning coverage knowledge point breadth and learning coverage knowledge point depth; evaluation scale V = (V) 1 ，v 2 ，v 3 ，v 4 ，v 5 ) Poor, qualified, good and excellent, respectively; after the service object completes the learning of the knowledge element, the system combines U and V to give a scoring matrix R; obtaining a fuzzy vector B = A × R through the weight vector A, and introducing a score vector S, so that the intelligence competitiveness I:
I＝B*S T
the weight vector and the score vector can be given by knowledge network experts or by an analytic hierarchy process.
(3) The self-expansion capability is measured by the comprehensive capability of the evaluation service object in the existing knowledge coverage range and the capability of expanding and solving the direct follow-up related knowledge element problem, and is mainly determined by the expansion degree of the expansion questions, the evaluation result of the expansion questions and the evaluation time;
(4) practical ability P ra The evaluation service object is obtained by evaluating the capability of the service object in solving the actual problem by using the obtained knowledge, and mainly depends on the solution score s, the problem difficulty d and the time t for solving the problem. The development problem capability or practice capability generated by a single development problem or practice problem is divided into:
wherein: k is a radical of 1 、k 2 The constant value is adjusted according to the actual situation, so that the result is more reasonable.
(5) Concentration force F f The learning active time of each learning is the ratio of the time that the service object concentrates on actually using for learning in the learning process to the total time spent learning knowledge points, i.e. the percentage of the actual effective learning time to the total learning time, measured by the learning active time ratio.
Wherein: t is t t As the distraction time, t is the total learning time. The distraction time consists of the page out-of-pause limit time, the page cut-out time and the page extra scroll time recorded in the service object full learning trace.
(6) The outline completion condition is measured by an outline requirement completion ratio, namely, the ratio of the completion degree of the outline after learning of a service object to the completion degree required by the outline as a whole after accumulating the products of the item-by-item completion degree of the outline and the importance degree of the outline is obtained by taking a knowledge element as a unit, and the outline completion condition of the knowledge element is calculated in the following way:
wherein: d i Is the importance of the ith outline, p i The completion degree of the ith outline service object is defined, and A is the completion degree required by the outline;
the above single self-expansion capability, single practice capability, single concentration force and single outline completion condition are obtained by adopting an iterative updating mode, wherein the overall expansion capability, practice capability, concentration force and outline completion condition is as follows:
wherein i = {1,2,3,4} represents self-expansion capability, practical capability, concentration force and outline completion condition respectively, and d i Numerical value, n, representing a single index i Representing the number of evaluations of a single index, T, at the start of an iteration i ＝0；
Overall multidimensional comprehensive learning evaluation:
E stu ＝(I，P，C d ，F，E，P ra )W T
wherein, I, P and C d 、F、E、P ra Respectively the intellectual competitiveness, the insights, the outline completion, the concentration and the self-relianceI extend ability and practice ability, W = (W) 1 ，w 2 ，w 3 ，w 4 ，w 5 ，w 6 ) For the weight, the weight W is obtained by an expert or by an analytic hierarchy process.
S4, recommending an individualized teaching scheme according to a teaching outline and multi-dimensional comprehensive learning evaluation of the service object;
the personalized teaching scheme refers to a learning guidance scheme which is recommended to a service object by the resource recommendation subsystem according to a teaching outline and multidimensional comprehensive learning evaluation acquired by the accurate online education system and comprises learning resources, a suggested learning path and the like, and the guidance scheme can enable the service object to achieve an ideal learning situation state according with the service object. And (3) the establishment of the personalized teaching scheme trains a prediction neural network model by using a large-scale data set to obtain a more accurate personalized education recommendation scheme. In this embodiment, the personalized teaching scheme is as shown in fig. 5, and the suggested learning path includes an evaluation method and content adopted by a knowledge subnet marked in the knowledge network to be covered at a certain target time.
S5, recording new learning traces and evaluation results generated by the service object learning according to the recommended personalized teaching scheme, and evaluating the personalized learning effect of the service object through a multi-dimensional comprehensive learning evaluation index;
the evaluation result is divided into an actual evaluation result and an expected evaluation result, wherein the actual evaluation result is calculated in the following way:
wherein G is i The score of the ith topic service object is represented.
The expected evaluation result is calculated in the following manner:
wherein s is i The score of the ith question; theta is the service object learning ability; p i (theta) the accuracy of the answer to the ith question of the service object, wherein the accuracy is calculated by the following method:
wherein: d is a constant, D =1.7, a is the discrimination of the test questions, b is the difficulty of the test questions, and c is a guessing parameter, which represents the probability that the service object can still answer the question when the comprehensive evaluation is negative infinity.
S6, judging the category of the service object according to the full learning trace and the multi-dimensional comprehensive learning evaluation index of the service object, providing a reinforced learning plan, expanding a personalized teaching scheme and optimizing a personalized learning promotion scheme;
the class to which the service object belongs is determined by using a fuzzy clustering analysis method and a fuzzy pattern recognition method.
According to six indexes of multi-dimensional comprehensive learning evaluation, through range or standard deviation transformation, the element values are normalized to an interval [0,1] and then multiplied by 100 to obtain the six-dimensional capability of the service object, and a six-dimensional tension graph of the service object is drawn. In this example, a six-dimensional tension chart is shown in fig. 6.
The service objects are divided into a plurality of types by a fuzzy clustering analysis method according to the relation among all indexes in the multi-dimensional comprehensive learning evaluation, and a system or a teacher can customize a teaching scheme according to the types of the service objects, so that a teacher and the system can conveniently recommend resources and adjust teaching according to the types of the service objects, and the types can be dynamically changed according to the full learning traces of the service objects. The fuzzy clustering analysis is shown in FIG. 7 and is described as follows:
selecting proper lambda threshold value by means of standardization, calibration, clustering and other methods, and dividing service objects into a plurality of types A = (A) 1 ，A 2 ，A i ，…，A n )，A i (i is more than or equal to 1 and less than or equal to n) respectively represents balanced type, perseveration type, intelligence type and the like.
Judging the service object by using a fuzzy pattern recognition method according to the type of the service object and the multidimensional comprehensive learning evaluation indexTo which category it belongs. Let classification scheme a = (a) 1 ，A 2 ，A i ，…，A n ) Service object multidimensional comprehensive learning evaluation index S = (S) 1 ，s 2 ，s 3 ，s 4 ，s 5 ，s 6 ) First, S and A are calculated i Proximity of (d):
wherein the content of the first and second substances,
then, the closest class is selected according to a closeness selection principle.
The comprehensive evaluation of the students after learning is taken as { intellectual competitiveness: 85, self-expanding ability: 80, practical ability: 73, retention force: 60, concentrated force: 71, outline completion: 65, for example, classify students into "high intelligence and perseverance".
The personalized learning promotion scheme refers to education resources such as recommended knowledge elements, test questions and data according to the difference between the individual learning effect of the service object and each index of the expected learning effect.
And S7, iteratively optimizing the personalized teaching scheme recommendation model by adopting a neural network model according to the learning trace and the multi-dimensional learning situation information generated by the service object.
The method specifically comprises the following steps: learning trace data and multi-dimensional learning situation information generated by all service objects of the system are fed back to the personalized teaching scheme recommendation subsystem and the clustering analysis recognition subsystem, and the neural network model is adopted to iteratively optimize the personalized recommended teaching scheme recommendation model, so that the learning scheme recommendation accuracy is improved, and the learning situation analysis capability is improved.
The iterative optimization personalized recommended teaching scheme recommendation model is based on the advantages of the recommended teaching scheme. The assessment of the goodness of the recommended teaching solution can be assessed using, but not limited to, the variance of the expected learning effect and the actual learning effect. In this example, teaching plan is recommendedThe goodness assessment is measured by the variance of the expected learning effect and the actual learning effect. Let column vector R = (R) 1 ，r 2 ，…，r n ) T Representing the actual learning effect of n service objects, column vector P = (P) 1 ，p 2 ，…，p n ) T Representing the expected learning effect of n service objects, let column vector C = R-P, calculate the expected E (C) of C, then the variance of the expected learning effect and the actual learning effect is:
D(C)＝E{[C-E(C)] 2 }
then, the superiority of the recommended teaching plan is judged according to the variance.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (7)
1. The multidimensional informatics analysis method based on the knowledge network accurate online education system is characterized by comprising the following steps:
constructing a knowledge network accurate online education system;
based on a knowledge network accurate online education system, recording multi-dimensional learning situation information according to learning traces of service objects;
forming multi-dimensional comprehensive learning evaluation of the service object by utilizing a knowledge network emotion comprehensive evaluation method based on the existing learning trace and multi-dimensional emotion information of the service object;
recommending an individualized teaching scheme according to a teaching outline and multi-dimensional comprehensive learning evaluation of a service object;
recording new learning traces and evaluation results generated by the service object learning according to the recommended personalized teaching scheme, and evaluating the personalized learning effect of the service object through a multi-dimensional comprehensive learning evaluation index;
judging the category of the service object according to the full learning trace and the multi-dimensional comprehensive learning evaluation index of the service object, proposing a reinforcement learning plan, expanding an individualized teaching scheme and optimizing an individualized learning promotion scheme;
according to learning traces and multi-dimensional learning situation information generated by the service object, a neural network model is adopted, and an individualized teaching scheme recommendation model is iteratively optimized;
various data generated when the service object uses the knowledge network accurate online education system to learn are recorded in the learning trace, wherein the data comprise learning time, learning duration, the breadth and depth of coverage of knowledge elements learned at a certain time, learning resource selection preference, learning behavior, learning speed and progress, evaluation arrangement, an evaluation method and evaluation result information;
the multidimensional comprehensive learning evaluation is evaluated through evaluation indexes, and the evaluation indexes comprise: outline completion, intelligence competitiveness, insights, concentration, self-expansion and practice;
the multidimensional comprehensive learning evaluation method comprises the following steps:
(1) the insights of the retention are measured by the time of study input, the time of study ending, the frequency of study input and the time of study input each time; the adherence force calculating method is that the variance with the current study input time, study end time, study frequency and study input time is calculated according to study input time, study end time, study frequency and study input time recorded in the study trace, and the variance vector D = (D) 1 ，d 2 ，d 3 ，d 4 ) A weight vector W = (W) from four elements 1 ，w 2 ，w 3 ，w 4 ) The final retention was obtained as:
(2) the intellectual competitiveness is measured by age, learning speed, evaluation effect, and the breadth and depth of learning coverage knowledge elements; u = (U) 1 ，u 2 ，u 3 ，u 4 ，u 5 ) Respectively indicate age, learning speed, evaluation effect and learning coverageThe breadth of the knowledge points and the depth of learning coverage of the knowledge points; evaluation scale V = (V) 1 ，v 2 ，v 3 ，v 4 ，v 5 ) Poor, qualified, good and excellent, respectively; after the service object finishes learning the knowledge element, the knowledge network accurate online education system combines U and V to give a scoring matrix R; obtaining a fuzzy vector B = A R through the weight vector A, and introducing a score vector S to obtain the intelligence competitiveness I:
I＝B*S T
wherein, the weight vector and the score vector are given by knowledge network experts or by an analytic hierarchy process;
(3) the self-expansion capability is measured by the comprehensive capability of the evaluation service object in the existing knowledge coverage range and the capability of expanding and solving the problem of directly subsequent related knowledge elements;
(4) practical ability P ra The evaluation service object is obtained by using the obtained knowledge to solve the actual problem, and the evaluation service object is divided into the following expansion problem capacity or practice capacity according to the solution score s, the problem difficulty d and the problem solving time t, wherein the expansion problem capacity or practice capacity generated by a single expansion problem or practice problem is as follows:
wherein: k is a radical of formula 1 、k 2 Is a constant;
(5) concentrated force F f As measured by the learning active time ratio:
wherein: t is t t Is the distraction time, t is the total learning time; the distraction time consists of the time of exceeding the pause limit of the page, the time of cutting out the page and the time of additionally rolling the page, which are recorded in the full learning trace of the service object;
(6) the outline completion condition is measured by an outline requirement completion ratio, namely, the ratio of the completion degree of the outline after learning of a service object to the completion degree required by the outline as a whole after accumulating the products of the item-by-item completion degree of the outline and the importance degree of the outline is obtained by taking a knowledge element as a unit, and the outline completion condition of the knowledge element is calculated in the following way:
wherein: d i Importance of the ith outline, p i Serving the completion degree of an object for the ith outline, wherein A is the completion degree required by the outline in whole;
the above single self-expansion capability, single practice capability, single concentration force and single outline completion condition are obtained by adopting an iterative updating mode, wherein the overall expansion capability, practice capability, concentration force and outline completion condition is as follows:
wherein i = {1,2,3,4} represents self-expansion capability, practical capability, concentration force and outline completion condition, p i Numerical value, n, representing a single index i Representing the number of evaluations of a single index, T, at the start of an iteration i ＝0；
And (3) overall multidimensional comprehensive learning evaluation:
E stu ＝(I，P，C d ，F，E，P ra )W T
wherein, I, P and C d 、F、E、P ra Respectively intelligence competitiveness, insistence, outline completion, concentration, self-expansion ability and practice ability, W = (W) 1 ，w 2 ，w 3 ，w 4 ，w 5 ，w 6 ) For the weight, the weight W is obtained by an expert or by an analytic hierarchy process.
2. The multidimensional informatics analysis method of claim 1, wherein the knowledge network precision online education system comprises:
(1) A knowledge network management engine;
(2) A knowledge network formed based on the correlation relationship of the knowledge elements;
(3) A knowledge network resource subsystem;
(4) A user management subsystem;
(5) A service object personalized evaluation subsystem;
(6) A schema management subsystem;
(7) A service object learning effect evaluation subsystem;
(8) A personalized teaching scheme recommendation subsystem;
(9) A clustering analysis and identification subsystem;
(10) And a learning context analysis subsystem.
3. The multidimensional informatics analysis method of claim 2, wherein the knowledge network management engine is responsible for the management of the knowledge elements, and the functions comprise: (1) adding, deleting and modifying knowledge points; (2) labeling a knowledge element; (3) storing the knowledge points and the knowledge elements; (4) constructing a knowledge network based on the relation of the knowledge elements; (5) establishing learning evaluation resources based on the knowledge network, the knowledge elements and the corresponding knowledge resources; (6) and recording and storing the learning trace.
4. The multidimensional informatics plot analysis method according to any one of claims 1 to 3, wherein the evaluation results are divided into actual evaluation results and expected evaluation results, wherein the actual evaluation results are calculated by:
wherein G is i A score representing the ith service object;
the expected evaluation result is calculated in the following manner:
wherein s is i The score of the ith question; theta is the service object learning ability; p i (theta) the accuracy of the answer to the ith question of the service object, wherein the accuracy is calculated by the following method:
wherein: d is constant, D =1.7; a is the discrimination of the test questions; b is the difficulty of the test question; c is a guess parameter which represents the probability that the service object can still answer the pair when the comprehensive evaluation of the service object is negative infinity.
5. The multidimensional informatics plot analysis method of any one of claims 1-3, wherein determining the class to which the service object belongs is given using fuzzy clustering analysis and fuzzy pattern recognition, comprising:
the service objects are divided into a plurality of types by a fuzzy clustering analysis method according to the relation among all indexes in the multi-dimensional comprehensive learning evaluation, and a system or a teacher can customize a teaching scheme according to the types of the service objects, so that a teacher and the system can conveniently recommend and adjust resources and teaching according to the types of the service objects, and the types can be dynamically changed according to the full learning traces of the service objects;
and judging the category of the service object by using a fuzzy pattern recognition method according to the type of the service object and the multi-dimensional comprehensive learning evaluation index.
6. The multidimensional informatics analysis method of claim 5, wherein the iterative optimization personalized teaching plan recommendation model is based on goodness of a recommended teaching plan, and the goodness evaluation of the recommended teaching plan is measured by variance of expected learning effect and actual learning effect.
7. A multidimensional informatics analysis method as claimed in any one of claims 1 to 3 and 6 wherein the service objects are students or learning organisations including classes, schools, learning groups and/or interest groups.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910073607.5A CN109858797B (en) | 2019-01-25 | 2019-01-25 | Multi-dimensional informatics analysis method based on knowledge network accurate online education system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910073607.5A CN109858797B (en) | 2019-01-25 | 2019-01-25 | Multi-dimensional informatics analysis method based on knowledge network accurate online education system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109858797A CN109858797A (en) | 2019-06-07 |
| CN109858797B true CN109858797B (en) | 2023-01-20 |
Family
ID=66896211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910073607.5A Active CN109858797B (en) | 2019-01-25 | 2019-01-25 | Multi-dimensional informatics analysis method based on knowledge network accurate online education system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109858797B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340660B (en) * | 2019-07-01 | 2023-09-01 | 黑龙江省华熵助晟网络科技有限公司 | Online learning auxiliary system and method |
| CN110473127A (en) * | 2019-08-21 | 2019-11-19 | 上海信立德教育科技集团有限公司 | A kind of teacher's levels method, apparatus, system and storage medium |
| CN112446009B (en) * | 2019-09-02 | 2022-06-21 | 北京新唐思创教育科技有限公司 | Learning path planning method and device, readable storage medium and server |
| CN112241447B (en) * | 2020-12-18 | 2021-04-02 | 北京世纪好未来教育科技有限公司 | Learning situation data processing method and device, computer equipment and storage medium |
| CN113516574A (en) * | 2021-07-27 | 2021-10-19 | 北京爱学习博乐教育科技有限公司 | Self-adaptive learning system based on big data and deep learning and construction method thereof |
| CN113486255B (en) * | 2021-09-08 | 2021-11-16 | 南京麦豆健康管理有限公司 | Internet-based postpartum online consultation matching system and method |
| CN114662995A (en) * | 2022-05-19 | 2022-06-24 | 山东经贸职业学院 | Online learning effect evaluation method and system based on artificial intelligence |
| CN116109456B (en) * | 2023-04-03 | 2023-07-28 | 成都大学 | Comprehensive evaluation method and system for intelligent education, electronic equipment and storage medium |
| CN116909339B (en) * | 2023-09-14 | 2024-01-09 | 厘壮信息科技(苏州)有限公司 | Intelligent household safety early warning method and system based on artificial intelligence |
| CN117114941B (en) * | 2023-10-25 | 2024-02-09 | 深圳市微校互联科技有限公司 | Education and teaching management platform based on dynamic learning condition data chain |
| CN117573985B (en) * | 2024-01-16 | 2024-04-05 | 四川航天职业技术学院(四川航天高级技工学校) | Information pushing method and system applied to intelligent online education system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104575142A (en) * | 2015-01-29 | 2015-04-29 | 上海开放大学 | Experiential digitalized multi-screen seamless cross-media interactive opening teaching laboratory |
| CN107103384A (en) * | 2017-04-01 | 2017-08-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of learner's study track quantization method based on three-dimensional knowledge network |
| CN108563749A (en) * | 2018-04-16 | 2018-09-21 | 中山大学 | On-line education system resource recommendation method based on various dimensions information and knowledge network |
| CN108573628A (en) * | 2018-04-23 | 2018-09-25 | 中山大学 | The method that H-NTLA based on study track is recommended with extension knowledge point set |
-
2019
- 2019-01-25 CN CN201910073607.5A patent/CN109858797B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104575142A (en) * | 2015-01-29 | 2015-04-29 | 上海开放大学 | Experiential digitalized multi-screen seamless cross-media interactive opening teaching laboratory |
| CN107103384A (en) * | 2017-04-01 | 2017-08-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of learner's study track quantization method based on three-dimensional knowledge network |
| CN108563749A (en) * | 2018-04-16 | 2018-09-21 | 中山大学 | On-line education system resource recommendation method based on various dimensions information and knowledge network |
| CN108573628A (en) * | 2018-04-23 | 2018-09-25 | 中山大学 | The method that H-NTLA based on study track is recommended with extension knowledge point set |
Non-Patent Citations (1)
| Title |
|---|
| 一种学习路径规划的模型及方法;陈霞等;《科研信息化技术与应用》;20171120(第06期);第5-14页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109858797A (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109858797B (en) | Multi-dimensional informatics analysis method based on knowledge network accurate online education system | |
| CN107230174B (en) | Online interactive learning system and method based on network | |
| Dutt et al. | Clustering algorithms applied in educational data mining | |
| CN110378818A (en) | Personalized exercise recommended method, system and medium based on difficulty | |
| CN112184500A (en) | Extraclass learning tutoring system based on deep learning and knowledge graph and implementation method | |
| Kuzilek et al. | Student success prediction using student exam behaviour | |
| Geetha et al. | Prediction of the academic performance of slow learners using efficient machine learning algorithm | |
| Nelyub et al. | Machine learning to identify key success indicators | |
| Ali et al. | Prediction MOOC’s for student by using machine learning methods | |
| Aji et al. | An implementation of C4. 5 classification algorithm to analyze student’s performance | |
| Atienza et al. | A Deep Neural Network in a Web-based Career Track Recommender System for Lower Secondary Education | |
| CN117313947A (en) | Learning early warning method based on random forest and tuna group optimization algorithm | |
| Fleischer et al. | Teaching and learning data-driven machine learning with educationally Designed Jupyter Notebooks | |
| Hussain et al. | Identifying beneficial sessions in an e-learning system using machine learning techniques | |
| Otero et al. | Finding informative code metrics under uncertainty for predicting the pass rate of online courses | |
| Siregar et al. | Analysis Of Decision Support Systems Edas Method In New Student Admission Selection | |
| Ruiz-Rodríguez et al. | Classification of student success using Random Forest and Neural Networks | |
| Carmen | Data analysis of short-term and long-term online activities in LMS | |
| Yurtkan et al. | Student Success Prediction Using Feedforward Neural Networks | |
| CN113822774A (en) | Financial teaching system, using method thereof and computer readable storage medium | |
| Emirtekin et al. | ONLINE COURSE SUCCESS PREDICTION OF STUDENTS WITH MACHINE LEARNING METHODS. | |
| Yauri et al. | A Machine Learning Approach in Predicting Student’s Academic Performance Using Artificial Neural Network | |
| Palani | Identifying at-risk students in virtual learning environment using clustering techniques | |
| Boumi | Development of a Multivariate Poisson Hidden Markov Model for Application in Educational Data Mining | |
| Kavya et al. | Student Personality Analysis In Blended Mode Using Big Five |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |